Escolar Documentos
Profissional Documentos
Cultura Documentos
(variability)
I. Introduction
II. Range
III. Average Deviation
IV. SS
V. Variance and Standard
Deviation
frequency frequency
5 6 7 8 9 10 11 12 5 6 7 8 9 10 11 12
Score Score
frequency frequency
5 6 7 8 9 10 11 12 5 6 7 8 9 10 11 12
Hand washings a day Hand washings a day
1
I. Introduction: Why Measure
Dispersion?
C. To make informed decisions:
Two routes to campus - same average
time
frequency frequency
5 6 7 8 9 10 11 12 5 6 7 8 9 10 11 12
Time (min) Time (min)
Measures of Dispersion
Range: describes the spread by giving
contrasting the two most extreme scores
Three ways to report it
1. Give both scores
“The grades ranged between 56 and 97.”
2. The highest score minus the lowest
score:
97-56 = 41
Note, to be accurate, you should use the upper real limit of the
high number and the lower real limit of the low number.
10.5 - 4.5 = 6
range
2
Measures of Dispersion
Range (continued)
3. High score - Low score + 1
If the scores are integers, adding 1 adds in the
upper and lower real limits.
Examples:
10-5+1 = 6
if 97.5 is the URL, and 55.5 the LRL
then 97.5-55.5 = 42
and 97-56+1 = 42
NOTE: This is the convention we will use in this
course!
Measures of Dispersion
Problems with the range:
1) Completely determined by extreme
scores.
2) Not very useful for describing the
shape of the distribution or drawing
conclusions.
Measures of Dispersion
Average deviation:
Average distance each score is from
the mean of the distribution.
x x-M |x-M|
8 8-3 = 5 5
1 1-3 = -2 2
3 3-3 = 0 0
0 0-3 = -3 3
∑= 12 0 10
3
Measures of Dispersion
Average Deviation (continued)
Conceptually very useful (closely
related to the standard deviation), but it
is unstable under sampling.
Measures of Dispersion
SS (sum of the squared deviations)
x x-M (x-M)2
8 8-3 = 5 25
1 1-3 = -2 4 SS = ∑(x-M)2
3 3-3 = 0 0 = 38
0 0-3 = -3 9
∑= 12 0 38
M = 12/4 = 3.0
Measures of Dispersion
Variance (σ2)
Population Variance is the average of
the squared deviations from the mean.
4
Measures of Dispersion
Standard Deviation (σ) for a population
Square root of variance
σ = √ σ2
= √∑(x-M
N
)2
= √ss/N
Example: √9.5 = 3.08
Measures of Dispersion
Summary
Range = high-low+1
SS = ∑(x-M)2
Population
Variance: σ 2 = SS/N
Measures of Dispersion
Example problem: Calculate the range,
SS, variance, and standard deviation
for the following populations of scores:
Group 1 Group 2
1 6
9 9
5 5
8 8
7 7
Range 9 5
Mean 6 7
SS
Var 8 2
Std 2.83 1.41
5
Measures of Dispersion
Visualizing what the standard deviation
measures.
- about halfway to the range
- the point of inflection on a frequency
graph
- most of the scores (68 %) should fall
within 1
standard deviation of the mean.
6
Measures of Dispersion
SS and the computational formula
Definition: SS = ∑(x-M)2
Examples:
7
Measures of Dispersion
Degrees of freedom and populations
estimates
Definition: number of pieces of
information that are free to vary.
Measures of Dispersion
Degrees of freedom and deviations
scores
2x -2x-M
df = 3
4 0
6 2
∑ = 10
M= 4
Measures of Dispersion
Because deviation scores are based on
the mean, once I have calculated the
mean I have reduced my degrees of
freedom by one. Thus, in sample
estimates of variance, the degrees of
freedom are n-1.
In general df = n - 1
8
Measures of Dispersion
Define:
s2 = sample variance = SS/n-1
s = sample standard deviation = √
SS/n-1
Measures of Dispersion
Summary
Range = high-low+1
SS = ∑(x-M) 2
Population Sample
Variance: σ 2=SS/N s 2 =SS/n-1
Measures of Dispersion
Sample Problems:
9
Measures of Dispersion
Factors Affecting Variability
1. Sample Size
a) Range increases with sample size
b) Variance decreases with sample size
2. Extreme scores - greatly influence
most measures, especially range
10